Modeling and interoperability of heterogeneous genomic big data for integrative processing and querying

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Next Generation Sequencing (NGS) generates vast amounts of diverse genomic and epigenomic data.
• Tertiary analysis, focused on understanding interactions between genomic regions and their products, is a significant challenge.
• Integrating and making sense of heterogeneous data formats is crucial for biological discovery.

Purpose of the Study:

• To introduce a paradigm shift in genomic data tertiary analysis.
• To present the Genomic Data Model (GDM) for unifying and describing diverse genomic data.
• To enable data interoperability and facilitate bio-molecular knowledge discovery.

Main Methods:

• Developed the Genomic Data Model (GDM) to link genomic features with experimental, biological, and clinical metadata.
• Mapped common genomic data formats (SAM, VCF, NARROWPEAK, BED) to the GDM.
• Integrated heterogeneous genomic datasets from multiple sources.
• Utilized the GenoMetric Query Language (GMQL) for querying integrated data.

Main Results:

• The GDM successfully integrates and homogeneously describes semantically heterogeneous genomic data.
• Demonstrated data interoperability through the GDM and GMQL.
• Enabled comprehensive processing of multiple, diverse data types.
• Provided a foundation for domain-specific computations and bio-molecular knowledge discovery.

Conclusions:

• The GDM offers a unified approach to managing and analyzing large-scale, heterogeneous genomic data.
• Combined use of GDM and GMQL enhances bio-molecular knowledge discovery.
• This model facilitates advanced tertiary analysis of (epi)genomic data generated by NGS technologies.